Empirical Rule in AP Statistics

In AP Statistics, the Empirical Rule (the 68-95-99.7 rule) says that in a normal distribution about 68% of values fall within 1 standard deviation of the mean, about 95% within 2, and about 99.7% within 3. It only applies to distributions that are approximately normal.

What is Empirical Rule?

The Empirical Rule, also called the 68-95-99.7 rule, is a quick way to describe where data lands in a normal distribution. If a distribution is normal with mean μ and standard deviation σ, then:

• About 68% of values fall between μ - σ and μ + σ.
• About 95% of values fall between μ - 2σ and μ + 2σ.
• About 99.7% of values fall between μ - 3σ and μ + 3σ.

Think of it as a ruler you lay on top of the bell curve, marked in standard deviations instead of inches. Because the normal curve is symmetric, you can split those percentages in half. For example, 95% within 2σ leaves 5% outside, so about 2.5% sits in each tail. Likewise, about 16% of values are above μ + σ, and only about 0.15% are above μ + 3σ.

There's an important catch. The rule is a property of the normal model. It does not work for strongly skewed or oddly shaped distributions, so you check the shape of the data (histogram, dotplot) before you use it. The percentages are also approximations, not exact values.

Why Empirical Rule matters in AP® Statistics

The Empirical Rule lives in Unit 1: Exploring One-Variable Data and Collecting Data, and Fiveable's practice questions tie it to Topic 1.10, The Investigative Question Revisited and Data Collection. The link to that topic shows up in the language of parameters. In the rule, μ and σ describe a whole population. Under AP Stats 1.10.B, a census records information from every individual in a population (EK 1.10.B.1), so a mean computed from a census is a parameter, not a statistic. That's why practice questions about the Empirical Rule often turn into "Is this a parameter?" questions.

The bigger reason it matters is that the rule teaches you to think in standard deviations. That habit carries through the whole course. You use it when you judge whether a value is unusual, when you work with probability distributions in Unit 2, and when you read the "about 2 standard errors" logic behind 95% confidence in Units 3 and 4.

How Empirical Rule connects across the course

Z-Score (Unit 1)

A z-score tells you how many standard deviations a value is from the mean. The Empirical Rule tells you what percentage of a normal distribution lies within z = ±1, ±2, and ±3. Put them together and you can instantly say that a value with z = 2.4 is in roughly the top 2.5%, which makes it pretty unusual.

Parameter (Units 1, 3, 4)

The μ and σ in the Empirical Rule are parameters because they describe an entire population. A census of every student's height gives you a parameter (AP Stats 1.10.B). A sample gives you a statistic that only estimates it. Keeping that straight is what Units 3 and 4 are built on.

Law of Large Numbers (Unit 2)

The 68-95-99.7 percentages are long-run proportions under a normal model. The Law of Large Numbers explains why real data from a truly normal process gets closer to those percentages as you collect more of it. A small sample might show 60% within 1 SD just by chance.

Confidence Intervals for Proportions and Means (Units 3-4)

When a sampling distribution is approximately normal, about 95% of sample results land within about 2 standard errors of the true parameter. That's the Empirical Rule applied to statistics instead of individual data values, and it's the intuition behind why a 95% interval looks like "estimate ± about 2 SE."

Is Empirical Rule on the AP® Statistics exam?

On multiple choice, the Empirical Rule usually shows up as a direct recall or quick-calculation stem. Typical examples are "What percentage of the data lies between μ - 3σ and μ + 3σ?" (about 99.7%) or "What percentage falls within 3 standard deviations of the mean?" Harder versions ask for a tail or a lopsided interval, like the percent between μ - σ and μ + 2σ (about 34% + 47.5% = 81.5%). Those questions reward you for drawing the curve and splitting percentages by symmetry.

You'll also see it paired with parameter questions. If heights from all students in a school average 67 inches, that average is a parameter, because it came from a census.

On free response, the skill is judgment, not memorization. Before you use 68-95-99.7, confirm the distribution is approximately normal. The 2018 FRQ Q5 gave histograms of teaching year at two high schools. That kind of data is exactly where you need to read shape, center, and spread from the graph rather than assume a bell curve. If a distribution is clearly skewed, saying "95% of teachers are within 2 SD of the mean" would be an unsupported claim.

Empirical Rule vs Z-Score (and exact normal calculations)

The Empirical Rule gives you approximate percentages, but only at exactly 1, 2, or 3 standard deviations from the mean. A z-score is the tool for any other distance. If a value is 1.7 SD above the mean, the Empirical Rule can only tell you the answer is somewhere between 68% and 95% territory. To get an actual proportion, you standardize to a z-score and use a calculator or table. Use the rule for quick estimates and sanity checks, and use z-scores for precise answers.

Key things to remember about Empirical Rule

• In a normal distribution, about 68% of values fall within 1 standard deviation of the mean, about 95% within 2, and about 99.7% within 3.

• The Empirical Rule only applies when a distribution is approximately normal, so check the shape of the data before using it.

• Because the normal curve is symmetric, you can halve the leftover percentage to get each tail, such as about 2.5% above μ + 2σ.

• The μ and σ in the rule are parameters because they describe a whole population, like a mean calculated from a census.

• The rule's percentages are approximations at whole standard deviations, so use z-scores and a calculator for exact proportions at other values.

• The same "about 95% within 2 standard deviations" idea shows up again with sampling distributions and 95% confidence intervals in Units 3 and 4.

Frequently asked questions about Empirical Rule

What is the Empirical Rule in AP Stats?

It's the 68-95-99.7 rule. In a normal distribution, about 68% of values lie within 1 standard deviation of the mean, about 95% within 2, and about 99.7% within 3.

Does the Empirical Rule work for any distribution?

No. It only works for distributions that are approximately normal (bell-shaped and symmetric). For a strongly skewed distribution, like many real histograms on FRQs, those percentages can be badly wrong, so check the shape first.

What percentage of data is within 3 standard deviations of the mean?

About 99.7% in a normal distribution, which leaves only about 0.3% outside, or roughly 0.15% in each tail. That's why values beyond 3 standard deviations are treated as very unusual.

How is the Empirical Rule different from a z-score?

A z-score measures how many standard deviations a single value is from the mean. The Empirical Rule tells you what percent of a normal distribution falls within 1, 2, or 3 of those standard deviations. For any z that isn't a whole number like 1, 2, or 3, you need a calculator or table instead of the rule.

Is the Empirical Rule on the AP Stats exam?

Yes. It appears in Unit 1 multiple-choice questions asking for percentages within 1, 2, or 3 standard deviations, and the idea behind it supports normal-based reasoning about sampling distributions and confidence intervals in Units 3 and 4.